Fraunhofer Institute for Integrated Circuits IISHow individual is disease, really – and what does that mean for its treatment? Personalized medicine addresses precisely this question: it regards biological diversity as the key and integrates molecular, imaging, and clinical data into individual disease profiles. Two central technologies in this field are spatial biology, which reveals the spatial distribution of biological molecules within tissues, and multi-omics, which combines analyses of different molecular layers such as genes, proteins, and metabolic processes. Researchers at Fraunhofer IIS have developed the software solution MIKAIA® to support this approach.
Personalized Medicine: From the Average to the Individual
Personalized Medicine: From the Average to the Individual
Juni 29, 2026 | Biological Diversity as the Key
Tissue sample (TMA core) from a patient with head and neck cancer (HNSCC) following PD L1 immunotherapy, analyzed using the MIKAIA® app “FL Cell Analysis.” Cell annotation and classification are shown on the right, while analysis results are displayed on the left as a cell scatter plot and marker statistics. Source: Max Delbrück Center.
Two people receive the same diagnosis, yet the treatment is effective for only one of them. Such differences reflect individual biological characteristics – and accordingly, diseases themselves may follow highly individual courses. Personalized medicine aims to systematically capture these differences and use them to guide more precise treatment decisions.
At Fraunhofer IIS, a key research focus of the Digital Health and Analytics department is the systematic integration of diverse data sources and data types. Imaging data are analyzed, contextualized within existing knowledge, and cross-referenced with the latest literature. The in-house developed, AI-powered image analysis software MIKAIA® supports researchers in systematically analyzing complex data and making them accessible.
This approach is particularly important in heterogeneous diseases such as cancer: tumors consist of diverse cell types that interact in complex ways. For the selection of effective therapies, it is often not sufficient to know which cells and molecules are present in the body – their spatial context is equally crucial.
When spatial information makes the difference
To better understand the principle of spatially resolved tissue analysis, a simple analogy can help: a supermarket. A traditional analysis – so-called bulk sequencing – shows which components are present overall, such as sugars, fats, or vitamins. Single-cell analyses go a step further, revealing which “products” are present and what they are made of – for example, high sugar and low vitamin content in sweets, or water and high vitamin content, as in fruit. However, the crucial distinction only emerges with spatial context: whether an item is located in the confectionery aisle or at the fruit stand.
The same applies in biology. It is not only the presence of cells and molecules that matters, but also their spatial arrangement within tissues. Spatial biology reveals precisely these structures – showing how cells are organized within tissues, which neighborhoods they form, and how they interact with one another. These spatial relationships are often crucial for disease progression and the success of therapies.
From complex data to actionable knowledge
Spatial biology unfolds its full potential in combination with multi-omics – the integrated analysis of different molecular layers, such as gene activity (genomics), protein distribution (proteomics), and metabolic processes (metabolomics). Only when combined do these approaches reveal how such processes interact and shape disease.
With the capabilities of spatial biology and multi-omics, data complexity continues to grow. Digital pathology produces high-dimensional, gigapixel-scale images that capture numerous markers, millions of cells, and their spatial relationships. The challenge is to translate these data into reliable and reproducible insights – turning relevant patterns into robust, evidence-based conclusions.
This is where the software solution MIKAIA® from Fraunhofer IIS comes in: it supports life science researchers in systematically analyzing spatially resolved omics data and helps make complex datasets comparable and accessible.
Spatial patterns as the key to better therapies
A concrete application of MIKAIA® is the analysis of immunotherapies in oncology: by comparing tissue samples, it becomes possible to identify how the spatial organization of immune and tumor cells differs between patients who respond to therapy and those who do not.
MIKAIA® enables the systematic identification of such spatial patterns and their comparison across multiple samples. This makes the biological mechanisms underlying therapy response or resistance visible – providing a key foundation for more precise, personalized treatment strategies.
The combination of spatial biology, multi-omics, and advanced data analytics highlights what matters most in the Medicine of the Future: understanding data in the right context and translating it into actionable knowledge that directly benefits patients.